Validation can be done at different stages of the process. Accordingly, there are three main types of validation as follows:
1. Prospective Validation – done before the process
commences
2. Concurrent Validation – done as the process is going on
3. Retrospective Validation – done on the already completed
process
Prospective Validation
It is defined as
establishing documented evidence that a given system does what it purports to
do based on a previously determined protocol. This type of validation is
generally carried out before the start of a new process of manufacture. It must
be done on a minimum of three consecutive batches of the product.
To carry out this validation, each step of the proposed
process is evaluated to determine which parameters are critical to the quality
of the finished product. With this information, experiments are designed and
documented in an authorized protocol.
Prospective validation protocol must cover the evaluation of
all the equipment, facilities, utilities, and analytical test procedures that
will be used in the production of the new product. Only after data has been
obtained about the critical process parameters, it will be possible to prepare
the Master Batch Records.
Using such a well-defined process, a series of product
batches must be produced. The number of batch runs to be carried out must be
sufficient to allow the collection of data for evaluation. Generally, three
consecutive batch runs are considered sufficient for a complete validation of
the process. However, in reality, more than three runs may also be required to
arrive at sufficiently reliable data.
During a validation run, the batch size must be kept the
same as that intended for regular industrial-scale production. If it is
intended to sell the validation batch products, care must be taken to produce
the batches in conditions that comply completely with cGMP (current Good
Manufacturing Practices). Also, such batches may be sold only after verifying
that the validation exercise has given a satisfactory outcome and been
authorized for marketing after passing all quality requirements.
When the validation batches are being processed, samples
should be drawn at frequent intervals and tests should be performed at
different stages of the production process; all results must be documented
thoroughly. Final products in their final packs must also be tested for
comprehensive data collection.
When deciding on the validation strategy, it is good to
obtain data using different lots of active ingredients and major additives.
Batches manufactured during different shifts, using different facilities and
equipment that will be used for commercial production, must be evaluated.
Readings must be taken over a wide operating range for the most critical
operations, and all data obtained must be exhaustively analyzed.
Once the data generated has been reviewed, guidelines can be
prepared regarding the level of monitoring necessary as a part of in-process
controls during regular production. All such guidelines should be made a part
of the Batch Manufacturing Record and Batch Packing Record. If necessary, they
must also be added to the relevant Standard Operating Procedures (SOPs).
Prospective validation data is also to be used to determine
limits, frequencies of testing, and actions to be taken in situations when the
limits are exceeded.
Information in Prospective Validation Protocol:
- Brief description of the process to be validated.
- Summary of the critical manufacturing steps to be studied.
- List of facilities and equipment to be used including monitoring/recording/measuring instruments/equipment and their calibration status.
- Analytical test methods to be used and their validation status.
- In-process controls proposed and their acceptance criteria.
- Sampling plan and procedures.
- Methods to record results and evaluate the data obtained.
- Specifications for finished product acceptance.
- Additional tests to be performed and their acceptance criteria and validation status.
- Proposed timeframe for the validation process.
- Functions and responsibilities in the validation program.
Concurrent Validation
Concurrent validation involves monitoring the critical
processing and testing steps at the in-process stage. It is almost the same as
prospective validation except that the manufacturer will sell the products
manufactured during the validation run, provided they meet all the
pre-determined quality requirements. There must be documents maintained that
show the justification for a concurrent validation, and due approval of the
decision by authorized persons. Documentation for concurrent validation is the same
as that for prospective validation.
Retrospective Validation
Retrospective validation is defined as establishing
documented evidence that a system performs as purported, by reviewing the
historical data that had been collected during the manufacturing and testing
stages. This validation is done for products that have already been
distributed; this method of validation is, therefore, acceptable, only for
processes that are well-established and stabilized over many years of
production. Retrospective validation is unsuitable in cases where there has
been any recent change in either the product composition, processing steps, or
equipment used in the manufacture and testing of the product.
A specific protocol must be prepared first, outlining how
the retrospective validation will be carried out. Historical data is collected
from batch manufacturing and packaging records, equipment logbooks, process
control charts, personnel change records, finished product testing data, and
stability test results. After historical data collection and review, results
must be reported, along with a conclusion and recommendations, if any.
Batches for retrospective validation must be selected in a
manner to represents all the batches made during the period selected for
review. The number of batches included in the validation must be sufficient to
prove the consistency of the process. Generally, data is collected from
anywhere between 10 and 30 consecutive batches. If fewer batches will be used,
the reason must be justified and documented. Any batches that did not meet the
specifications during the review period, must also be included. In some cases,
samples retained after distribution may be tested to obtain necessary data.
Elements to be considered for retrospective validation:
- Batches manufactured during the defined period (ex – 10 last successive batches).
- Batches are released per year.
- Batch size and strength.
- Master manufacturing and packaging records.
- Latest specifications for APIs and finished product.
- Process deviations list.
- Corrective actions list.
- Records of manufacturing document changes/revisions.
- Stability testing data for several batches.